The prevalence of obesity has increased dramatically over the last three decades, as has its corresponding burden of disease. Much of that burden is preventable. But while there is general agreement about the role of economic, physical, and social environments in shaping health behaviors, it is unclear how suggested policies perform in terms of effectiveness and costs across countries. There is no shortage of suggestions how to address the obesity epidemic - the gap is how to assess the likely long-run effects of interventions and how that varies by country or population. As quality and quantity of data vary across countries and the ultimate outcomes of interest (prevention of chronic illness at the population level) will not manifest for many years following an intervention, mathematical models like the one we develop in this project are needed to integrate the information from multiple sources. This project develops a microsimulation model (MSM) to analyze and predict consequences of obesity prevention policies across countries and to provide an integrated decision support tool. The specific aims of this project are to: (1) Develop, test and validate a dynamic microsimulation modeling platform, capturing both individual and social dynamics of diet, physical activity, and BMI for the US, Australia, Mexico, and 4 other countries; (2) Use the platform to project the health, social and economic consequences associated with current behavioral risk factors; (3) Assess the outcomes and costs of policies in multiple areas (food taxes and subsidies, labeling laws, active transportation, media campaigns, voluntary industry efforts/self-regulation) at reducing the social harms associated with inactivity and excess caloric intake; (4) Provide an open internet version of the model to researchers and policy makers to conduct their own calculations.